chore(cli): be consistent between workers and expose ExtraLLamaCPPArgs to both (#3428)

* chore(cli): be consistent between workers and expose ExtraLLamaCPPArgs to both

Fixes: https://github.com/mudler/LocalAI/issues/3427

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* bump grpcio

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

backend/python/autogptq/requirements.txt

@@ -1,6 +1,6 @@
 accelerate
 auto-gptq==0.7.1
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi
 transformers

backend/python/bark/requirements.txt

@@ -1,4 +1,4 @@
 bark==0.1.5
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi

backend/python/common/template/requirements.txt

@@ -1,2 +1,2 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf

backend/python/coqui/requirements.txt

@@ -1,4 +1,4 @@
 TTS==0.22.0
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi

backend/python/diffusers/requirements.txt

@@ -1,5 +1,5 @@
 setuptools
-grpcio==1.66.0
+grpcio==1.66.1
 pillow
 protobuf
 certifi

backend/python/exllama/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi
 setuptools

backend/python/exllama2/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi
 wheel

backend/python/mamba/requirements.txt

@@ -1,3 +1,3 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi

backend/python/openvoice/requirements-intel.txt

@@ -2,7 +2,7 @@
 intel-extension-for-pytorch
 torch
 optimum[openvino]
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 librosa==0.9.1
 faster-whisper==1.0.3

backend/python/openvoice/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 librosa
 faster-whisper

backend/python/parler-tts/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi
 llvmlite==0.43.0

backend/python/rerankers/requirements.txt

@@ -1,3 +1,3 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi

backend/python/sentencetransformers/requirements.txt

@@ -1,3 +1,3 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi

backend/python/transformers-musicgen/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 scipy==1.14.0
 certifi

backend/python/transformers/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi
 setuptools==69.5.1 # https://github.com/mudler/LocalAI/issues/2406

backend/python/vall-e-x/requirements.txt

@@ -1,3 +1,3 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi

backend/python/vllm/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.66.0
+grpcio==1.66.1
 protobuf
 certifi
 setuptools

core/cli/worker/worker.go

@@ -2,6 +2,7 @@ package worker
 
 type WorkerFlags struct {
 	BackendAssetsPath string `env:"LOCALAI_BACKEND_ASSETS_PATH,BACKEND_ASSETS_PATH" type:"path" default:"/tmp/localai/backend_data" help:"Path used to extract libraries that are required by some of the backends in runtime" group:"storage"`
+	ExtraLLamaCPPArgs string `name:"llama-cpp-args" env:"LOCALAI_EXTRA_LLAMA_CPP_ARGS,EXTRA_LLAMA_CPP_ARGS" help:"Extra arguments to pass to llama-cpp-rpc-server"`
 }
 
 type Worker struct {

core/cli/worker/worker_llamacpp.go

@@ -3,6 +3,7 @@ package worker
 import (
 	"fmt"
 	"os"
+	"strings"
 	"syscall"
 
 	cliContext "github.com/mudler/LocalAI/core/cli/context"
@@ -12,7 +13,6 @@ import (
 )
 
 type LLamaCPP struct {
-	Args        []string `arg:"" optional:"" name:"models" help:"Model configuration URLs to load"`
 	WorkerFlags `embed:""`
 }
 
@@ -34,9 +34,8 @@ func (r *LLamaCPP) Run(ctx *cliContext.Context) error {
 		"llama-cpp-rpc-server",
 	)
 
-	args := os.Args[4:]
+	args := strings.Split(r.ExtraLLamaCPPArgs, " ")
 	args, grpcProcess = library.LoadLDSO(r.BackendAssetsPath, args, grpcProcess)
-
 	args = append([]string{grpcProcess}, args...)
 	return syscall.Exec(
 		grpcProcess,

core/cli/worker/worker_p2p.go

@@ -25,7 +25,6 @@ type P2P struct {
 	NoRunner           bool   `env:"LOCALAI_NO_RUNNER,NO_RUNNER" help:"Do not start the llama-cpp-rpc-server"`
 	RunnerAddress      string `env:"LOCALAI_RUNNER_ADDRESS,RUNNER_ADDRESS" help:"Address of the llama-cpp-rpc-server"`
 	RunnerPort         string `env:"LOCALAI_RUNNER_PORT,RUNNER_PORT" help:"Port of the llama-cpp-rpc-server"`
-	ExtraLLamaCPPArgs  string `name:"llama-cpp-args" env:"LOCALAI_EXTRA_LLAMA_CPP_ARGS,EXTRA_LLAMA_CPP_ARGS" help:"Extra arguments to pass to llama-cpp-rpc-server"`
 	Peer2PeerNetworkID string `env:"LOCALAI_P2P_NETWORK_ID,P2P_NETWORK_ID" help:"Network ID for P2P mode, can be set arbitrarly by the user for grouping a set of instances" group:"p2p"`
 }
 

docs/content/docs/features/distributed_inferencing.md

@@ -68,7 +68,7 @@ And navigate the WebUI to the "Swarm" section to see the instructions to connect
 To start workers for distributing the computational load, run:
 
 ```bash
-local-ai worker llama-cpp-rpc <listening_address> <listening_port>
+local-ai worker llama-cpp-rpc --llama-cpp-args="-H <listening_address> -p <listening_port> -m <memory>" 
 ```
 
 And you can specify the address of the workers when starting LocalAI with the `LLAMACPP_GRPC_SERVERS` environment variable:
@@ -98,7 +98,7 @@ To reuse the same token later, restart the server with `--p2ptoken` or `P2P_TOKE
 2. Start the workers. Copy the `local-ai` binary to other hosts and run as many workers as needed using the token:
 
 ```bash
-TOKEN=XXX ./local-ai worker p2p-llama-cpp-rpc
+TOKEN=XXX ./local-ai worker p2p-llama-cpp-rpc --llama-cpp-args="-m <memory>" 
 # 1:06AM INF loading environment variables from file envFile=.env
 # 1:06AM INF Setting logging to info
 # {"level":"INFO","time":"2024-05-19T01:06:01.794+0200","caller":"config/config.go:288","message":"connmanager disabled\n"}